v1.0 GDW 2004-May-03
Background
During the April 2004 DEIMOS run, summit staff reported a problem inserting a filter into the filter wheel. Investigation revealed that the filter wheel had sustained damage and that several filters were subsequently scratched.Findings
Filter wheel
We found that the filter wheel had been damaged at slot 6. The "left" (dewar) side of the filter wheel (as viewed from the load position) was found by summit staff to be bent inward toward the filters, making it difficult to insert filters and apparently causing the edge of the filter wheel to scrape across the glass surface of the filters during changeouts. Staff immediately took remedial action, bending the filter wheel away from the filters in order to prevent further damage to the optics. Figure 1 shows the state of the filter wheel following the intervention; the filter wheel is now bows out away from the filter.Optics
Three filters show significant scratches on the glass surface (along the side facing the dewar) resulting from scraping against a piece of metal which protruded into slot 6 on the wheel. The worst scratches are on the NB8580 filter (the new narrowband interference filter recently purchased by the megamask teams) and the BAL12 filter (not a major concern, since GG400 can be used instead for most programs) The V filter is also scratched, slightly less than the others. Observable effects resulting from this damage have not been determined.Photos
Figure 1: Photo showing the bent part of the DEIMOS science filter wheel as seen from the filter load position to the side of DEIMOS. The green arrows indicate the bent portion of the filter wheel; note that this section is not parallel to the edge of the filter holder as designed. This filter assembly drawing shows the wheel as seen from the front of the instrument; item 1 on this diagram shows the wheel (note the relative thinness of the wheel at the centers of the filters).
Figure 2:
Photo of the DEIMOS NB8580 filter, which apparently sustained
the worst damage. A series of scratches (note region between
arrows) runs the length of the filter, parallel to the direction
of insertion.
Figure 3:
Photo of the DEIMOS BAL12 filter, with scratches the length of
the filter as indicated by the arrows.
Figure 4:
Photo of the DEIMOS V-band filter, which apparently sustained
the least damage. The center of the filter is significantly
scratched and smaller scratches extend to the edge of the
filter (see arrows).
Conclusions
It seems certain that the filters were damaged during insertion and/or removal from slot 6 on the DEIMOS filter wheel because part of the aluminum filter wheel was deformed and protruded into the path of the optic. We do not yet have any good ideas as to how the filter wheel got warped in such a way as to scrape these filters. It is highly unfortunate that 3 filters were damaged before we learned of the problem. To avoid putting any more filters at risk in this slot, we moved the R-band filter (which is used for daily focusing and hence is never removed) into slot 6.
Gregory D. Wirth <wirth@keck.hawaii.edu> Last modified: Thu Oct 13 14:05:09 HST